M
MiniMax M2.1OPEN
MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development.
Prices update automatically — checked daily against provider list prices.
- $0.52 per 1M tokens, assuming 3 input for 1 output the convention used by Artificial Analysis
- cheaper than 36% of the 191 open-weight models tracked here
- 9 of 11 MiniMax models tracked here, from cheapest
Benchmarks & performance
Independent benchmark scores for MiniMax M2.1, measured by Artificial Analysis. Higher is better.
For the same budget, MiniMax M2.1 delivers 9.7 times more capability. Open weights also mean you can self-host it and pay nothing per token.
Models at this level cost $0.72 per 1M tokens (median of 17) — this one costs $0.52.
Cheaper and better on this index: GLM 5.3 Flash · Gemini 3.7 Flash (batch) · GPT-5.6 Luna and 13 more
About this model
MiniMax M2.1 is an open-weight AI model by MiniMax. You can download and self-host it for free; the prices below are hosted-API list prices, tracked daily, for when you prefer convenience over self-hosting.
Frequently asked questions
What is MiniMax M2.1?
MiniMax M2.1 is an AI language model from MiniMax. It is open-weight: you can download it and run it on your own hardware, for free. It scores 32.1 on the Artificial Analysis intelligence index.
Is MiniMax M2.1 free?
The weights are free and open — you can self-host MiniMax M2.1 and pay nothing per token. If you prefer a hosted API, list prices are $0.3 per million input tokens and $1.2 per million output tokens.
What is MiniMax M2.1 good at?
Independent benchmarks from Artificial Analysis give it GPQA 83%, MMLU-Pro 87.5%, Humanity's Last Exam 23.2%, Long Context Reasoning 66.3%, LiveCodeBench 81%, SciCode 40.7%, AIME 2025 82.7%, IFBench 69.9%, τ²-Bench 85.4%, Terminal-Bench Hard 28.8%. It is particularly used for mathematical reasoning.
Can I self-host MiniMax M2.1?
Yes. MiniMax M2.1 has open weights, so you can download it and run it on your own GPU or server with tools like Ollama, vLLM or llama.cpp — with no per-token cost.